While you’re here, you may wish to attend one of our upcoming training courses:
When we ask complainants in a workplace investigation whether there were any witnesses to the events that form the basis of their allegations, it is not uncommon to hear, “Well no, but I told my partner/best friend/colleague everything.” This is especially true in cases of sexual harassment or assault, where the events in question often take place in private, without witnesses present.
We refer to witnesses who have been told about the allegations but have not actually witnessed anything firsthand as confidant witnesses, and in the case law, their evidence is referred to as a prior consistent statement. It can be tricky to know what to do with this type of evidence. Should you speak to the partner, the best friend, the colleague? If you do, how should you weigh their evidence? At first glance, this evidence can appear highly relevant: if the complainant told her best friend the same story she tells you, doesn’t that consistency make it more likely she’s telling the truth? After all, if the complainant told a different story to her friend, that inconsistency would affect your credibility analysis.
The problem with this logic is, as the Human Rights Tribunal of Ontario Adjudicator put it in Ornelas v. Casamici Restaurant (2010 HRTO 1078), “The likelihood that the story is true is not increased because it has been repeated many times.” A complainant (or a respondent) can tell a consistently false story. Take, for example, the evidence that was before the Adjudicator in Ornelas. The applicant alleged that she was fired after she confronted the respondent about sexual comments that he made towards her, as well as an incident of sexual touching. In support of her allegations, the applicant pointed to the fact that, after her employment was terminated, she told the respondent, the respondent’s brother-in-law and her doctor that she was fired as a result of sexual harassment. The Adjudicator declined to rely on this evidence.
On the other hand, in some cases, a complainant would have to go to great lengths to create a trail of prior consistent statements, making it more likely that those statements support the complainant’s credibility. For example, consider a hypothetical complainant who alleges she was sexually assaulted by a co-worker following after-work drinks. This complainant tells a friend the next day, discloses to her partner six weeks later and files a formal complaint two months after that. You interview her friend and partner, and both report a story consistent with the one the complainant gave you. If this complainant was lying about the sexual assault, she would also have to have lied consistently to her friend and partner or roped them into her deception.
To illustrate this point further, contrast the situation in Ornelas with that in Morgan v. University of Waterloo (2013 HRTO 1644), another Human Rights Tribunal of Ontario application. In Morgan, the applicant alleged unwanted sexual touching by the respondent at a conference dinner and dance. The applicant told a colleague about the respondent’s actions the next day and the Adjudicator relied on this evidence as supporting the applicant’s story.
Ornelas and Morgan demonstrate the significance of timing and a motive to fabricate when weighing the evidence of confidant witnesses. In both cases, the respondents argued that the applicants were lying. However, in Ornelas, the applicant told people that she was fired as a result of sexual harassment after she was fired, days and weeks after the alleged incidents of harassment. In contrast, in Morgan, the applicant told a colleague about the unwanted sexual touching almost contemporaneously; the day after the incident occurred. The applicant’s prior consistent statements in Ornelas were less credible because they were made after the applicant had a motive to lie about why she was fired.
As an Adjudicator put it in a British Columbia Human Rights Tribunal application, in order to rely on a prior consistent statement made by a complainant or respondent to a confidant witness, the statement “must have been made before the speaker had a motive to fabricate his or her evidence.”
Whether or not to rely on the evidence of a confidant witness and what weight this evidence should be given is a decision that must be made on a case-by-case basis, with regard to the particular facts of the investigation. The following questions can assist you in making this determination:
1. Do you really need the evidence? If you have two eyewitnesses to the respondent kissing the complainant on the dancefloor at the holiday party, do you really need to interview the complainant’s roommate, who heard about the incident afterwards? Given the pitfalls associated with confidant witnesses, sometimes it’s best to avoid them altogether.
2. If you do really need the evidence, consider the timing. When did the complainant disclose the allegations to someone else? Keeping in mind that a complainant (or respondent) can tell a consistently false story, a disclosure made before the complaint was made and the investigation began will generally hold more weight than one made after, when the party has an obvious motive to lie.
3. Is there any motive to fabricate? Consider whether the complainant or respondent had any reason to lie to the confidant witness. Consider as well whether the party might have misperceived or misunderstood events and reported that misunderstanding to a confidant. For instance, a complainant may genuinely believe that the respondent was talking about him behind his back and told a co-worker about this when in fact you know that the respondent was talking about someone else.
A final note on confidant witnesses: parties often do not realize that confidant witnesses are, in fact, witnesses since they were not present for the events in question. So, it is always wise to ask the parties whether they told anyone about the events in question around the time those events took place, in order to determine if any confidant witnesses exist.
 Pashtoukh v. JR Hair Design and Russel, 1999 BCHRT 43 at para. 75.